(1) Field of the Invention
The present invention relates to a traffic measuring apparatus and method, and particularly to a traffic measuring apparatus and method for collecting information related to traffic caused by call control within an exchange.
(2) Description of the Related Art
Generally, various kinds of exchanges such as a station exchange are designed such that when the exchange provides call control for terminals accommodated in the exchange, a central controller (CC) of the exchange records within a main memory (MM) information on calls for the accommodated terminals over a predetermined period of time. Based on the call information, the central controller collects and creates various kinds of traffic information such as basic traffic information related to the activity ratio of the central controller, route traffic information, subscriber traffic information, and service traffic information as needed, and outputs the traffic information to a maintenance engineer (maintenance terminal) for the exchange at predetermined intervals (e.g., 15-minute intervals).
The traffic information is utilized as a reference for future design of exchanges, improvements of operation conditions, and measures against the occurrence of abnormality and congestion. For example, based on route traffic information, trunks are added/removed, restrictions are placed on alternative routes, and like actions are taken; based on the subscriber traffic information, new services to be provided for the subscribers or the like are determined; and based on the service traffic information, services are expanded/abolished and like actions are taken.
At present, in each exchange the above-described collection and creation of traffic information (traffic measurement) is performed through use of a call-by-call scheme or a sampling scheme. In the call-by-call scheme, traffic information is collected with reference to call information in the memory device whenever a call is generated for the accommodated terminals. In the sampling scheme, traffic information is collected with reference to all call information at constant intervals without regard to the generation of calls.
However, in traffic measurement utilizing the call-by-call scheme, traffic information is collected upon generation of each call. Therefore, although the measurement results (traffic measurement values) are extremely accurate and reliable, the load of the traffic measurement process itself increases in proportion to the number of calls. Accordingly, if the running level (the priority imparted to execution of a traffic measurement program) is set to the same level as the running level of call processing, there arises a chance that call processing, which must be performed preferentially, is adversely affected, resulting in an increase in the loss probability, for example.
Therefore, in order to reduce the influence of the traffic measurement processing on call processing, the running level of the traffic measurement processing is generally set to be lower than that of the call processing. In this case, if a heavy load state of the call processing continues, traffic measurement processing is maintained in a waiting state for a prolonged period of time, with the result that traffic measurement may not be completed within a predetermined period of time (e.g., within 15 minutes). In this case, the central controller outputs to a maintenance terminal the collected information at a time after the call processing load decreases to a predetermined level (e.g., after one hour).
That is, in the call-by-call scheme, when call processing is in a heavy load state, traffic measurement requires a very long time to complete, with the result that measurement results cannot be obtained at short intervals (e.g., 15-minute intervals).
Meanwhile, in traffic measurement employing the sampling scheme, since traffic information is collected at constant intervals without regard to the generation of calls, the traffic measurement processing load is always constant and low regardless of the number of calls, so that measurement can be completed within a short period of time (15 minutes) even when call processing is in the high load state. However, since traffic information is thinned down in terms of time for collection thereof, the overall reliability and precision of the measurement results become lower than those in the call-by-call scheme.
Since the traffic measurement performed in conventional exchanges fixedly utilizes either the call-by-call scheme or the sampling scheme but not both, the traffic measurement can be performed while importance is attached to either securing high reliability of results of traffic measurement or shortening the traffic measurement period, but not both. Therefore, detailed traffic measurement cannot be performed flexibly.